Magnetism is very strange

quote:

Originally posted by Rasher:

where did the earth get its magnetic field from?

Eddies in the space/time continuum.

(He'd probably know the answer.)

Okay AD. That article implies that all crystal structures and atoms are magnetic and that we just deal with the alignment in order to create an object with a big enough number of magnetic poles in a single direction. I suppose it figures that all crystals have a magnetic direction as they were formed on earth in a magnetic field, but it still doesn't explain why magnetism occurs, although it does explain the space manual alignment scenario.

Some links:
Types of Magnetism
Although they don't cover the 6th type of magnetism, superparamagnetism. This is what makes audio tapes, etc. possible.
And the home page here.
BTW magnetic fields DO exist in space. You don't need an atmosphere or other "media" to support magnetic fields.
Facinating stuff.
Here's an interesting question: If you have a bucket of nails and you use a permanent magnet to pick them up a few at a time by introducing the magnet into the bucket, attracting the nails, removing the magnet, pulling the nails off the magnet and placing them in, say, another bucket, how is it that you don't quickly wear out the permanent magnet?
Discuss.
David

Superparamagnetism.
Hmmm. Wasn't that one of the songs in Mary Poppins?

Can't you strike some metals with a hammer and they take on a level of magnetic property ??

Seem to remember something like that from my days in metalwork at school.

I can remember changing the magnetic properties by repeatedly hammering a steel needle during physics lessons at school and watching how the needle would point in a slightly different direction when suspended freely above the desk; the same could be observed on a much larger scale with ships due to the millions and millions of hammer and rivet blows suffered in the shipyard while under construction. This resulting magnetic field had to then be mapped and corrected so that in a pre GPS era the ship's compass could be relied upon to0 point to magnetic north.

Spacecraft may surf the solar system on magnetic fields.

quote:

Originally posted by Roy T:

the same could be observed on a much larger scale with ships due to the millions and millions of hammer and rivet blows suffered in the shipyard while under construction. This resulting magnetic field had to then be mapped and corrected so that in a pre GPS era the ship's compass could be relied upon to0 point to magnetic north.

I'm pretty sure there are still a couple of large poles poking out of Auckland Harbour that were once used to line up ships so that their compasses could be adjusted - by attaching tiny magnets to the side of the compass iirc.

quote:

Originally posted by Deane F:

quote:

Originally posted by Roy T:

the same could be observed on a much larger scale with ships due to the millions and millions of hammer and rivet blows suffered in the shipyard while under construction. This resulting magnetic field had to then be mapped and corrected so that in a pre GPS era the ship's compass could be relied upon to0 point to magnetic north.

I'm pretty sure there are still a couple of large poles poking out of Auckland Harbour that were once used to line up ships so that their compasses could be adjusted - by attaching tiny magnets to the side of the compass iirc.

Think this is called 'swinging the compass' and the boat (or aeroplane, etc.) has to be rotated 360 deg to produce a deviation chart (or adjustment of magnets around the compass) so that the correct heading can be taken, compensating for the effects of metal bits in the boat/plane.

You can also modify the magnetic field by 'degaussing' said boat/plane/etc.

/dl

Here is a very cool way of understanding magnetism - it is actually just relativity!


To understand this we need to remember several simple things

• like electrical charges repel and unlike charges attract e.g. two electrons will push away from each other.
  
  
• materials are made of negatively charged electrons and positively charged ionic cores. In a metal the electrons
  
  flow (current) when a voltage is applied, but the positive ionic cores are essentially fixed in the metal lattice.
  
  
• Not so simple - Special relativity says that when we view a moving object its length contracts relative to the
  
  length one would measure when it is stationary. i.e. if a spaceship were passing is at 90% of the speed of light we
  
  would see it shorter than when it was parked beside us. This is one of relativity's weirdnesses - which comes because
  
  the speed of light is the same for observers in all inertial frames. Google is your friend if you want to explore
  
  that further.
  
  
• Electrons moving in a wire do so at about 1% of the speed of light.
  
  
• There are lots of electrons in a wire.
  
  
• the wire is locally neutral - in other words at any point on the wire the postive and negative charges cancel.
  
  

So now we try to understant the attractive magnetic force beween two current carrying parallel wires. Let me call

them the left wire and the right wire and take the case that the currents are flowing in the same direction.

Remember what we have in each wire then are electrons thundering along the wire while the ionic cores are

stationary. However seen from the perspective of an electron in one wire (moving at ~1% of light speed), we have

ionic cores in both wires moving backwards and a static electron sea in the other wire.

Now we remember our relativity. As viewed by an electron in the left wire the distance between positive ions in the

right wire will be contracted (just like the spaceship) so the effect will be to increase the positive charge

density in the right wire.

The electron density in the right wire is unaffected because it looks stationary to the electron in the left wire.
The net effect then is that the electron in the left wire sees a net positive charge in the right wire and is thus

attracted to it - just as it is to any positive charge. This electrostatic attraction is EXACTLY magnetism.

Cool or what!

So magnetism can be explained as nothing other that the fact that because of relativity, a net charge imbalance is

induced and this causes elecrostatic attraction or repulsion! Electricity and Magnetism are REALLY facets of the

same thing. We just percieve them as different because our everyday experience doesn't involve things moving at

near-light speed.



Materials which are magnets in the normal 'horseshoe' magnet sense are exactly the same except now the 'currents'

are actually internal to the atoms and can be thought of as electons spinning around the atom in an orbit. The

alignment of all the spinning axes of the individual atoms is what is produced in the kind of cooling effect

mentioned above where a particular orientation is frozen into the material.

I hope this is intelligible!

Ian

PS An exercise for the reader is to use the above arguments to explain why parallel wires with currents flowing in

opposite directions repel on another. No new physics is needed.

Interesting explanation. Doesn't seem to be completely in line with other explanations I've seen. Do you have some sources for this argument? I'd enjoy seeing more detail.
If I follow your reasoning, EVERY magnet is an electromagnet then? Not sure I agree with that.
BTW, the electricity flowing in wires is far faster than 1% of the speed of light. Typical transmission line velocity factors are 60 to 70% and I've measured the speed of propagation in such lines many times. It is always a much larger fraction of the speed of light than 1%.
David

David,

I'll try and dig out some references. I've always fancied doing the calcs for the parallel wires to reproduce the standard em results but never found the time. Maybe I should, I just took it on trust from a book called 'Thinking Physics is Gedanken Physics' by Epstein. There is not much more detail in there than what I've given.

My understanding is that at the microscopic level every magnetic field is generated by the movement of charge, so in that sense is an electromagnet. The simple setup I used above is not exactly the same since a current circulation for example round an atom is not an inertial frame (the electron accelerates by changing its direction) However is gets a bit subtle when you think for example of the intrinsic spin of an electron, which has no positive core - yet has a finite magentic moment.

The 1% speed I quoted was a typical Fermi velocity in copper, I agree in tranmission lines the signals are transmitted much faster, but I didn't want to open up the whole phase vs group velocity issue. I was really just trying to illustrate the cool idea.

Let me know if you come across any good material on your searches.

Ian

My electronics and physics are less adavanced than yours dear Ian, but surely if by transmission lines you mean the National Gride, as the current is alternating [oscilating at 50 Htz] then effectively over time the electrons are making no actual displacement change at all except over the distance they repeatedly oscilate in ech cycle? Is that right? I have never dared to ask the question for fear of looking a fool till now. As I have now worked that out for myself, I no longer worry about asking the question!

ATB from Fredrik

And I thought this was a thread about Gordon Brown's charisma.

All negatively charged in that case I am sure! Fredrik

quote:

Originally posted by Fredrik_Fiske:
My electronics and physics are less adavanced than yours dear Ian, but surely if by transmission lines you mean the National Gride, as the current is alternating [oscilating at 50 Htz] then effectively over time the electrons are making no actual displacement change at all except over the distance they repeatedly oscilate in ech cycle? Is that right? I have never dared to ask the question for fear of looking a fool till now. As I have now worked that out for myself, I no longer worry about asking the question!

ATB from Fredrik

Fredrik,

I think the transmission lines David is referring to are more specialised usually high frequency structures. (But he'll correct me if I'm wrong - maybe he's a power engineer!) Usually in a wire the voltage is the same all along the wire. But if the time over which the voltage is changing is shorter than the time the signal takes to propagate to the end of the wire this is no longer true and then the detailed geometry of the wire (often called a waveguide in this case) is crucial in detemining the impedance.

Normal co-ax cable in a TV ariel is one such example - that is why you can actually get a signal into your TV without the co-ax core touching the socket - the EM wave propagating down the coax propagates across the gap and into the coax on the TV side.(provided it is close enough of course)

Your 50Hz question is exactly as you described it - the electrons in the wire wiggle back and forth in response to the field, not changing their average position.

There are no silly questions - only silly answers.

best wishes

Ian

Aha! I need not have worried about looking idiotic then! Fredrik

Ah...
The propagation/migration of the electrons is about 1% of C whereas the actual voltage/current that appears is much faster.
Is that what you mean?
When you have a couple of hundred meters of coax (with ready access to both ends such as on a spool), a really good signal generator and 2 channel 'scope, watching the input pulse and then then "waiting" for the pulse to come out the other end of the coax is quite interesting. The signal is certainly traveling at ~70% of C. But I don't have a way to observe the electrons!
And yes, I was speaking about transmission lines such as coax for RF and audio cables for the 70% velocity factor.
This is facinating stuff for sure.
And there are no silly questions on this topic, the person that can answer some of the mysteries is surely in line for a Nobel prize or two!

Just remember: Magnetic force is considered to be a "weak" force... The strong ones hold particles together!

David

quote:

Originally posted by NaimDropper:
Ah...
The propagation/migration of the electrons is about 1% of C whereas the actual voltage/current that appears is much faster.
Is that what you mean?

David

Yes, it is the real velocity of the electrons which is relevant in the relativistic argument.

Ian

Thank you, Ian.
Now I've got to do some digging in my old text books...
David

Hello Ian,

I'm not sure that Fitzgerald contraction really explains the attraction between the parallel wires. I'd put it down to the fact that a force is exerted on a current carrying conductor in a magnetic field.

But on the subject of relativity. I am reminded of the only interesting lecture in three years of boredom. It was our last Field Theory lecture.

The lecturer told us not to take notes but to enjoy the lecture, he thought that we would enjoy this. (Seemed unlikely). Anyway after years of struggling with Maxwells equations and all sorts of confusion, he set out to show that magnetism doesn't exist as a separate force.

We started out with the usual electrostatic forces between charged particles, he then considered the case when the charges are moving.

So far so good and not so much fun.

He then considered what would happen if they were moving very fast, say close to light speed.

He then applied the usual relativistic correction factor which had already bored us to tears earlier in some other series of lectures.

At this point things started to get interesting. I was quite keen to see where this was going.

Anyway a load of equations further, what did we end up with but the Maxwell's equations.

So he managed to demonstrate that what we call magnetism is really a relativistic effect caused by the fact that electricity flows (not the electron drift velocity) at close to light speed.

I wish I could repeat the calculation, but, apart from a short time working as a nuclear physicist, it's not the sort of thing I've ever needed to bother with.

Thought you might be interested.

Regards,

Paul.

hi Paul,

I'm not quite sure how to answer your post - we're actually agreeing but you don't notice it

If you take your second half first about deriving Maxwell's equations from Lorenz invariance and electrostatics this is another, more general, way of saying what I tried to say. In fact you can also begin with Lorenz invariance and magneto-statics and 'derive' electrostatics. Either way it is clear that the electric and magnetic fields are two facets of the same phenomenon.

Fitzgerald contraction IS a simple consequence of Lorenz invariance and so if you dig into the derivation of Maxwell's equations you were shown, you'll soon see that the source terms (charge densites) which arise in the derivation are 'enhanced' by the Fitzerald contractions in just the same way that I described. It was perhaps just not explicitly drawn out as such in the derivation - but that is the underlying physics.

Either way, is is to me one of the most beautiful and elegant things in physics.

cheers

Ian

The fact that electromagnetism can be derived from coulomb's very simple law of electrostatic forces
+ special relativity, and nothing more, is rather fascinating, and many people appear to rediscover it independently now and then.
There exists a book that derives the whole lot, including radiation. I've seen it. I've held it. I don't remember the name but it was from 1968

A central example derives the same result by using electromagnetism on one hand,
and special relativity on the other.
You have two charges which exert a coulomb force on each other.
If you now make them move forward , slowly, it doesn't have to be fast, in parallel, then electormagnetism says they exert an extra
magnetic force on on each other.
If you choose instead to leave the charges alone and you move backwards, slowly,
then you should apply a relativistic transformation on the forces between the electrons.
The result should be the same. As a next step you can sit near one charge while the other charge is moving.

The reason you see this relativistic magnetism effect with simple wires and slow moving currents is because
electrostatic forces are so huge, they just cancel out most of the time so you don't notice.

Imagine what this all adds up to. The mechanism doesn't just work for electrostatic forces.
All that relativity has to work with is a simple force with
some meaningless measure Q of one object, another measure q from the other object,
multiplied with the inverse of the square of the distance between the objects.

So we can imagine making this little diddy, a maxwell-einsteinian-transmogrifier.
On one end you put in a formula of forces between static objects that you made up or that maybe matches something real.
and out at the other end comes the maxwellian-sortakinda version of the thing,
with formulas describing magnetism-like behaviour and radiation-like behaviour.

Version 0.1 just does the generic inverse square power forces(which is special in its own way).
Version 1.0 will do whatever formula for a static radial force.
And version 2.0 will accept er, whatever.

This is going to be great. So you take the next force available, gravitation.
Instead of the force going qQ/rr it goes mM/rr which looks juss poifik. Instant gravito-magnetism.
And it doesn't work.

The problem is that when you move backwards from the two masses, the masses also transform.
Well then the transmogrified version will just look a bit different but roughly the same innit?
But the masses are a huge problem. You can't make the formulas consistent.
If you calculate one way you'll get different results than when you calculate another way, which is nonsense.
You can't just put any formula in the transmogrifier, there are some limitations, and gravity is just squatting right on top of one.

I imagine that Einstein must have spent a lot of thought on gravity before releasing his special relativity.
He had to be certain that he could claim special relativity was not broken from the start just because it did not apply to gravity.
(smirking peer reviewer: "oh we try the second force we can think of and you say 'it doesn't work there, Newton's law of gravity must be broken'?")
Then after publishing special relativity he had to find a way to take the mass out of that equation(other than saying there must be two kinds of mass).
And he did. Which was nice.

I still wonder, wouldn't it be possible to make some kind of approximate model of gravitation that looks like gravito-magnetism.
Probably yes. Approximations are funny that way. They can be good or bad, not just right or wrong.